Pattern of in-patient neurologic review: An experience from a Tertiary Hospital North-Western Nigeria.

BACKGROUND: Increase in neurologic diseases burden has increased the demand for neurology services globally, despite the shortcomings of shortage and maldistribution of neurologists worldwide, including Sub-Saharan Africa. This has placed significant pressure on the few available neurologists to provide optimal and effective services in our resource-challenged settings.
METHODS: Neurology referrals were prospectively reviewed over a period of 3 months. Sociodemographic characteristics of the patients, the initial diagnosis by the requesting team, the reasons for the consult/referral, the requesting personnel, duration of hospital stay before request, the time interval between receiving the request and review, the working and final diagnosis after the review, and the diagnostic outcome of neurologic review were analyzed using Predictive Analytics Software® version 18.0.0 for Windows (Chicago, Illinois, USA).
RESULTS: Fifty-three hand written in-patient requests were reviewed over the period of study given an average rate of 4.4/week. The mean age was 50.8 ± 16.1 years, and the median length of stay before a review was 1 day (interquartile range: 1-2.5 days). Diagnostic outcome of the reviews were; new diagnoses in 4 (11.3%), the incorrect diagnosis changed in 8 (15.1%), contribution to the differential diagnosis in 15 (28.3%), and contribution to management plans in 24 (45.3%) patients. The association between diagnostic outcomes and mortality in our study was not statistically significant (χ2 = 6.66, P = 0.08).
CONCLUSION: Our study showed that in-patient neurologic reviews led to significant improvement in diagnostic and management plans. Appropriate policy guidelines should focus more on efficient ways of maximizing benefits of these reviews to patients without overburdening the few available neurologists in our environment.

Introduction

Report by the World Health Organization showed 8 out of 10 disorders in the three highest disability classes are neurologic problems.[1] Annual estimates of up to 1 billion people have a neurology related disease constituting approximately about 6.3% of all disease burden, which is projected to increase by 12% over the next 20 years.[2] Reports from Nigeria have also shown that neurologic diseases constitute a significant burden to medical admissions, which was comparable to studies done elsewhere.[3456]

The specialist is expected to provide consultative advice or to perform tests in a consultation, while he/she is expected to share or assume total responsibility for specific interventions or on-going management of the referred health problem in a referral.[7]

By the generalist system of care in most of our health institutions, patients are admitted into the wards by nonneurologists, which may, therefore, prompt a neurology referral or consultation. While the burden of neurological diseases is increasing globally, shortage, and maldistribution of neurology specialists also exist even in resource challenged settings, and, therefore, significant pressure is placed on the available few for the provision of optimal, effective, and adequate care.[8910]

This may stem from the perceived belief that neurology is a difficult specialty, and current health care regulations and financing policies, especially in developed countries, favor mainly the procedural specialties over less procedural specialties.[1112]

The paucity of data in our environment served as the background for our study in assessing the frequency, sources, reasons, and outcome of these reviews.

Methods

The study was conducted in an urban tertiary referral health facility located in Kano. It is a 600 bedded hospital and presently has 14 clinical departments. The medical wards have a bed capacity of 80 for both male and female patients and an average of 5 to 10 new in-patient admissions daily.[13] It has a multidisciplinary diagnostic center for cardiology, gastroenterology, pulmonology, and neurology subspecialties. During the period of study, the neurology unit comprises two consultants and two specialist registrars who assist in the management of adult neurology patients from a wide range of medical, surgical, and allied specialties.

The study was a cross-sectional survey that lasted for 3 months (September 1 to November 31, 2014). During the study period, one of the two consultant neurologists reviewed the hand written requests received, after an initial review by the specialist registrar. Relevant information extracted from the questionnaire included; patient demographic characteristics, the initial working diagnosis by the requesting team, the reason for request, the requesting physician, the time interval between receiving the request and review (waiting time), length of hospital stay (LOS) before request, the working and final diagnosis after the neurological review, and the outcome of neurologic review.

The extracted data were analyzed using PASW® Statistics version 18.0 (SPSS Inc. Chicago, Illinois USA). All informal (verbal) and curbside referrals and consultations were excluded from the study.

Definition of terms

New neurological diagnosis – the neurologic consultation makes a new/novel neurological diagnosis where no previous diagnosis existed

Incorrect diagnosis changed – the neurologic consultation changes an incorrect diagnosis made by the referring physician

Contribution to differential diagnosis – the neurologic consultation suggests an additional differential diagnosis (ses) and/or additional test (s) to narrow down the differential diagnosis

Contribution to management plans – the neurologic consultation suggests a treatment plan for the neurological condition

No definite contribution – the neurologic consultation had no impact on patient management.

Categorical data were described as proportions or percentages, and comparisons made using the Chi-square test. Quantitative data were described as means and standard deviation, or median and interquartile range (IQR) where the data are skewed. At 95% confidence interval, P < 0.05 was considered significant.

Approval was obtained from Institutional Ethics Committee certified by the Nigerian National Health Research Ethics Committee before commencement of the study. The provisions of the Helsinki declaration were respected throughout the study.

Results

Overall 53 hand written in-patient requests were reviewed during the 3 months of study with a consultation/referral rate of 4.4/week. Referral (full responsibility for patients’ management) was the main reason for the request in 37/53 (69.8%) patients while shared care; and a second opinion were requested for in the remaining 9/53 (17.0%) and 7/53 (13.2%) patients, respectively. The mean age of the patients was 50.8 ± 16.1 years, of which 24/53 (45.3%) were female, and 39/53 (78%) reside in an urban setting. Importantly, about half of the patients in the study were not fully employed 23/53 (43.4%), and were predominantly full-time homemakers. The median LOS for the patients before a consult was requested was 1 day (IQR: 1–2.5 days).

Table 1 shows the time interval to a review for all consultations and referrals received. Review of 35/41 (77.4%) patients was done promptly (within 12 h) after they were received usually.

Table 1

Time interval to a review for all consultations and referrals received

Time (h)	Request-received interval	Received-review interval
<12	41 (77.4)	35 (66.0)
12-24	10 (18.9)	14 (26.4)
24-48	2 (3.8)	3 (5.7)
>48	-	1 (1.9)

The sources of neurology requests and their diagnostic categories are detailed in Table 2, which showed that 41/53 (77.4%) were received from the medical wards while the remaining 12/53 (22.7%) were from nonmedical wards (accident and emergency, orthopedics and maxillofacial surgery). Direct request by the specialty consultant during ward rounds was seen in 27/53 (50.9%), while the remaining 26/53 (49.1%) were requested by the specialty registrar, usually during calls.

Table 2

Diagnosis and sources of in-patient neurology requests

Diagnosis	Specialty										Total
	Cardiology	Endocrinology	Gastroenterology	Hematology	Infectious disease unit	Maxillofacial surgery	Nephrology	Orthopedics	Pulmonology	Rheumatology
Stroke/transient ischemic attack	6	6	9	4	3	-	3	-	1	1	33
Seizures	-	-	1	-	2	-	1	1	-	-	5
Central nervous system infection	-	-	-	-	-	1	-	-	1	-	2
Movement disorder	-	-	-	-	-	-	1	-	-	-	1
Intracranial space-occupying lesion	1	-	-	-	-	-	-	-	-	-	1
Myelopathy	-	-	-	-	-	-	4	-	2	1	7
Neuropathy	-	-	-	-	-	-	1	-	-	-	1
Uncertain	-	-	-	-	-	-	1	-	-	-	1
Others	-	-	-	-	-	-	1	-	1	-	2
Total	7	6	10	4	5	1	12	1	5	2	53

Stroke/transient ischemic attack was the most common diagnostic category, accounting for 33/53 (62.3%) of cases reviewed, which was followed by myeloradiculopathy and seizures/epilepsy in 10/53 (13.2%) and 5/53 (9.4%) cases respectively. One patient who had neuroimaging evidence suggestive of an ependymoma, and another who had nonorganic psychosis, were both allotted to the “other” category group and were referred for further evaluation to neurosurgery and psychiatry respectively. The only patient assigned to the “uncertain” category had a diagnosis of multiple cranial nerve palsies and could not afford magnetic resonance imaging because of its prohibitive cost and available only in a commercial diagnostic setting outside our center.

Table 3 shows the contribution of the review process relative to the outcome. New or novel diagnoses were made in 6/53 (11.3%) patients, and the diagnosis was changed in 8/53 (15.1%) of the patients. Contributions to the differential diagnosis and management plans were also made in 15/53 (28.3%) and 24/53 (45.3%) of cases, respectively. However, one patient died before any further investigation could reveal a definitive diagnosis. Of the patients that were alive 7/37 (18.9%) showed no clinical improvement while 2/37 (5.4%) were referred for surgical evaluation and the remaining 28/53 (75.7%) who showed clinical improvement, were discharged to see clinic for onward follow-up. Our study also showed that the association between diagnostic contributions and mortality was almost statistically significant (χ2 = 6.66, P = 0.08).

Table 3

Contribution of the diagnostic category in relation to outcome

Diagnostic category	n (%)	Alive n (%)	Died n (%)	χ2, P
New neurological diagnosis	6 (11.3)	3 (8.1)	3 (18.8)	χ2=6.66,P=0.08
Incorrect diagnosis changed	8 (15.1)	6 (16.2)	2 (12.5)
Contribution to differential diagnosis	15 (28.3)	14 (37.8)	1 (6.3)
Contribution to management plans	24 (45.3)	14 (37.8)	10 (62.5)
No definite contribution	0 (0.0)	0 (0.0)	0 (0.0)
Total	53	37	16

Discussion

The average consultation and/or referral rate of 4.4/week in our study could be extrapolated to 228.8/year, which is similar to a previous study reported by Adebayo et al. in South-Western Nigeria, suggesting that our data did not show any significant regional difference in terms of frequency for neurological diseases within Nigeria.[14] The seemingly low rates may partially be explained by changing seasonal patterns of patient admission and possibly by the high cost of care in tertiary health facilities in Nigeria, considering the out of the pocket system of health financing by patients.

The main reason for neurologic requests was for assuming full responsibility of patients’ management, which was similar to reports from healthcare facilities elsewhere in Nigeria and even in developed countries with specialized neurology units.[15] However, this being the only reason does not completely explain sometimes the more defensive medical reasons given, were requests may be driven by obvious mortality concerns.

Stroke accounted for most of the in-patient neurology referrals, which may be a reflection of the high incidence of stroke in our environment, as previously reported by Owolabi et al.[3] Practice policy of involving the appropriate specialty inpatient care from admission in our center may partly account for the increased frequency of requests to the neurologist. In addition, requests may also be driven by foreseeable logistical difficulties in making a definitive neurologic diagnosis using the appropriate specialized techniques, sometimes leading to the occasional “wait and see” approach by nonneurologist with its less than optimal consequences.

Confidence in the expertise of neurology team in the care of stroke patients by colleagues, despite the absence of a dedicated stroke unit in our institution, may have additionally accounted for the high requests seen in our study. However, data from a previous survey suggest no change in the pattern of referrals seen despite the presence of a specialized stroke unit in their hospital.[16]

The recent trend in the senior residency training by both postgraduate medical colleges in Nigeria of using the Calman pathway which is structured to focus more on their specialty, may limit their experience to neurology related cases, and probably contribute to the high neurology requests seen.[1517] Our study showed that the patients benefited from these reviews, comparable to similar studies done elsewhere that showed significant change in case diagnosis and management,[1819] shorter average length of stay,[20] and less likelihood of dying during admission when seen by a neurologist, even though conducted in entirely different patient populations.[141821]

This may also serve as a prelude for further analyzing the cost-effectiveness of the neurologist in our resource challenged environment, which hitherto has been presumed to be “invisible.”[16]

The outcomes from our study may be a clarion call to improve the neurology workforce right from undergraduate medical education, in addition to providing a robust continuing medical education of the general physicians on common neurology related case scenarios and management protocols/guidelines. Furthermore, special emphasis to neurology during general medicine rotation in residency training, which will enhance and improve utilization of neurologic services in our resource, challenged settings.

Our findings should be interpreted within the limitations of being a cross-sectional survey in a tertiary hospital setting, where presumably no definite inferences on the cause and effect should be drawn, as our study subjects probably comprises more advanced neurologic cases who were not randomized.

Conclusion

Our study has shown that in-patient neurology consultations and referrals have led to significant improvement in-patient diagnosis and management plans. However, appropriate resources management and allocation, especially in terms of work force and training in our environment is needed to further improve services and avoid overburdening the few available neurologists. More prospective studies are required to examine the critical impact of referrals and consultations in other specialties, with a view of improving them within the context of our resource challenged setting.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.